Designed around tokens/sec benchmarks, cost-per-GB tracking, and compatibility matrices across NVIDIA, Apple Silicon, and AMD -- for people who run models, not rent them.
Illustrative launch-preview data. Live price tracking is not connected yet.
Launching 2026. Inference benchmarks, cost tracking, compatibility data.
Illustrative price signals / launch preview
What We Track
Which GPU runs your model, how fast, and at what cost. Benchmarks across platforms, new and used pricing, and the setup guides to get running.
Planned capability
Real inference speed on real models. We test every GPU across Llama, DeepSeek, Mistral, and more -- with llama.cpp, MLX, and vLLM backends. Quantized and full precision.
Planned capability
New retail, eBay used, Apple refurb. We track cost-per-GB of usable memory across every inference-capable GPU and unified memory config. Updated daily.
Planned capability
Can your GPU run it? Enter a model name and size. We show every GPU that fits it, the expected tokens/sec at each quantization level, and what it costs.
Planned capability
Mac + ollama. NVIDIA + llama.cpp. AMD + ROCm. Step-by-step from unboxing to first inference, with benchmarked settings for your exact hardware.
Planned capability
Check live availability and pricing at the providers we track: Rent on Vast.ai ☆ · Rent on Amazon Web Services · Rent on Microsoft Azure · Rent on Baseten · Rent on Civo · Rent on CoreWeave · Rent on Crusoe · Rent on Cudo Compute · Rent on Denvr Dataworks · Rent on DigitalOcean · Rent on fal · Rent on GPU.ai · Rent on Hot Aisle · Rent on Hyperstack · Rent on Jarvis Labs · Rent on Koyeb · Rent on Lambda · Rent on Latitude.sh · Rent on LeaderGPU · Rent on Lium · Rent on Massed Compute · Rent on Modal · Rent on Nebius · Rent on Oracle Cloud Infrastructure · Rent on OVHcloud · Rent on Replicate · Rent on Runpod · Rent on SaladCloud · Rent on Scaleway · Rent on Seeweb · Rent on Spheron · Rent on Thunder Compute · Rent on Together AI · Rent on Verda · Rent on Voltage Park
Provider links marked with ☆ may earn us referral credit at no cost to you.
This launch preview maps the inference benchmarks, cost trackers, and compatibility data planned for the full product.